We propose to continue and extend the analysis of blood samples that were collected at baseline as part of the ongoing Physicians' Health Study (PHS), a randomized trial of aspirin and beta-carotene among 22,071 U.S. male physicians aged 40-84 years. The aspirin component of the trial was terminated in 1988 when a 44% reduction in myocardial infarction was observed in the aspirin group, but the beta-carotene component was recently extended for an additional five years. Blood samples were collected prior to randomization from 14,916 participants free from prior NH, stroke, or cancer (excluding nonmelanoma skin cancer), and were stored at -82 degrees C. Using a nested case-control design, we plan to continue to analyze the specimens for retinol, beta-carotene, other carotenoids, vitamin E, seleniura, lipids, and fatty acids. It is likely that any chemopreventive effect of beta-carotene will be modified by baseline blood levels, reflecting intake from dietary sources. If the anti-cancer activity of beta-carotene operates chiefly or solely among those with initially low blood levels, the trial may not have sufficient power to detect this effect, because that target group would not be identifiable. However, the proposed analyses will provide the opportunity to observe that effect. The primary aim of this project is therefore to enhance the power and interpretability of the PHS findings. For maximum utilization of the samples, we will also test a number of hypotheses relating blood levels of various nutrients listed above to risk of subsequent cancer, myocardial infarction, and stroke. We will analyze specimens for cases and controls individually matched for age and smoking. Several of these nutrients may have important interactions with beta-carotene as well as among themselves. We also plan to extend the analyses relating lipid fractions to myocardial infarction and stroke, and test several new hypotheses: homocysteine (and related nutrients, vitamin B12 and folate), and plasminogen activator inhibitor in relation to MI risk, and iron stores and risk of cancer. Several important and intriguing results have already emerged from the study. The nested case-control approach and the multiple analyses provide an efficient means to increase the scientific value of the PHS and test a number of hypotheses of potential public health importance, using blood samples that were collected at considerable expense.